Project Summary The National Research Council and Institute of Medicine documented that compared to other developed countries, two-thirds of US male disadvantage in life expectancy occurs before age 50. This project will examine mortality risk among an understudied and hard-to-reach but exceptionally high-risk subpopulation: gang members. Gangs are street- and youth-oriented groups that share a collective identity and engage in high levels of criminal activity. Over 8% of the population has a history of gang involvement by their early 20s, and gang members are disproportionately male, black or Latino, and low SES?groups at high-risk for early mortality. Prior studies have reported mortality statistics on gang members, but this work suffers from small sample sizes of gang members and deaths, sometimes crudely constructed comparison groups (or none at all), and a lack of systematic data collection on mortality, including cause-of-death. We propose to conduct the largest and most comprehensive investigation into the gang membership-mortality association extant. Derived from law enforcement gang intelligence gathered between 1993 and 2003, we have a database composed of 3,154 male gang members from the St. Louis metropolitan area (mean age=20, 90% black). We recently linked these data with the National Death Index (NDI) through 2015. NDI records indicate that there are 1,900 possible death matches, of which 106 were determined to be ?exact match deaths? (e.g., name and date of death identical) and a total of 262 are ?assumed dead? (name and date approximate but not identical). These values likely constitute the lower (3.4%) and upper (8.4%) bounds of mortality risk. Our next step will be to assess the veracity of matches with other searches and data sources. While prior studies have concentrated on homicide, our theoretical framework leads us to focus broadly on external causes of death: accidents, suicide, and homicide, as well as gun- and drug-related deaths. We will develop baseline estimates of overall and cause-specific mortality risk and calculate years of potential life lost among gang members. We will examine factors associated with mortality, including factors exogenous to gangs and factors endogenous to gangs, while also accounting for temporal variation in mortality risk over 23 years. Our project is novel in that it will (1) link a non-traditional data source with the NDI, offering an opportunity to examine the quality of matches in this hard-to-reach population, (2) provide the most comprehensive investigation into the gang member-mortality link, (3) serve as a model for future NDI research with high-risk subpopulations, and (4) yield critical information about an important yet overlooked explanation for the life expectancy disadvantage among US males. National health goals and social policies that consider the mortality risk of gang members will gain added insight into an important high-risk population and a crucial yet overlooked explanation for the comparatively low US life expectancy.